The presence of voids behind tunnel linings can have a substantial impact on the safety performance of the tunnel. An accurate and quick evaluation for the tunnel safety performance is of significant importance for the maintenance of the infrastructure. In the current study, the minimum relative safety coefficient is proposed to evaluate the safety performance of the lining, and an artificial neural network is applied to predict the safety performance of tunnel with void behind lining. Firstly, the numerical simulations verified by physical model tests were employed to examine how the void defects affect the safety performance of the lining. Key findings indicate that void defects instigate substantial alterations in the distribution of inner force. The dangerous points are primarily concentrated in the void and its vicinity or the arch foot. Furthermore, the numerical simulation test scheme is generated using an orthogonal test design, thereby avoiding the omission of extreme working conditions that pose a danger. Finally, an artificial neural network model is proposed to predict the safety performance based on the results of numerical simulation. Nine factors are considered as input variables including the location of void, the circumferential length of void, the radial length of void, the longitudinal length of void, the buried depth of tunnel, the lateral pressure coefficient, the surrounding rock class, the lining concrete class, and the deterioration degree of lining, while the minimum relative safety coefficient is considered as the output variable. The accuracy of the prediction model could reach up to 97.53%, illustrating its effectiveness of evaluating the safety performance of defected tunnels.
扫码关注我们
求助内容:
应助结果提醒方式:
